Error In Cels With Individual Nonmem
with individual 15, (ID number 16). Weighted sum of squared individuals residuals is infinite. Message issured from estimation step at initial obj function evaluation. To generate these index plots I used ADVAN3 TRANS4. In an effort to determine that the error was not due to the data input, I added one patient at a time and ran the model. I was able to generate estimates up until this patient. What error in the code or data would generate this error message? Thank you MK Grandison ***** Date: Mon, 14 May 2001 15:37:03 -0700 (PDT) From: stuart@c255.ucsf.edu The actual error message is: WEIGHTED SUM OF "SQUARED" INDIVIDUAL RESIDUALS IS INFINITE It is to be understood pretty much at face value. An individual residual is the difference between a response and its individual-specific prediction, (when there are L2 records, the response may be multivariate, and the residual will be a vector difference). A squared individual residual is the square of the residual (when the residual is a vector; it's square is the scalar product of the vector with itself). The phrase "weighted sum" may be confusing. It actually refers to the sum of weighted squared individual residuals. A weighted squared individual residual is the squared residual multiplied by the reciprocal of the intraindividual variance of the residual (when the residual is a vector, the weight is the inverse of the intraindividual variance-covariance of the residual). The sum in question is that sum of weighted squared residuals which ordinarily shows up in the expression for the conditional likelihood for the data from the individual. When can it be infinite? Here are a few possibilities: 1. There is an error in the coding for a prediction, so that the prediction is very large (in absolute value), and along with a very small intraindividual variance, the weighted squared residual is effectively infinite. (E.g. the prediction is a logarithm, whose argument is mistakenly coded so that it is very small.) 2. The prediction is very large due to an incorrect value on the data record containing the response. 3. An initial estimate of an element of SIGMA is mistakenly very small, so that an intraindividual variance is very small, and the weighted squared residual is effectively infinite. 4. A prediction is mistakenly very small and this has the effect of making the intraindividual variance very small (as with a constant cv intraindividual variance model), and so the weighted squared residual is effectively infinite. 5. A intraindividual variance model is being used which legitimately produces a very small varianc
] [ by thread ] [ by subject ] [ by author ] [ by messages with attachments ] From: Mark Sale - Next Level Solutions Date: Mon, 13 Feb 2012 14:56:29 -0700 Yuhong, = ; You likely have a predicted value of 0 at some point (maybe a predo= se sample?). With a predicted value of 0, the proportional error vari= ance will be 0, and you have the additive error variance fixed to 0. = So, http://www.cognigencorp.com/nonmem/nm/99may112001.html the total error variance will be zero. Since the "weight" for the= sum of squares is the error variance, and this appears in the denominator = of the calculation of the sum of weighted squared deviations, you're gettin= g an infinite sum of squares. You might consider not fixing the addit= ive error variance to 0.Mark= Mark Sale MDPresident, Next Level http://www.cognigencorp.com/nonmem/current/2012-February/3110.html So= lutions, LLCwww.NextLevelSoln= s.com 919-846-9185A carbon-neutral companySee= our real time solar energy production at:http://enlighten.enphaseen= ergy.com/public/systems/aSDz2458 -------- Original Message -------- Subject= : [NMusers] NONLINEAR PK MODEL ERROR From: Yuhong Chen = Date: Mon, February 13, 2012 4:20 pm To: nmusers_at_globomaxnm.com Dear nmusers, I have a compound wi= th nonlinear CL. I tried to run the nonlinear model but got error message a= s following, could someone point out what could be the error in my control = file or my data? Best regards, = Yuhong 0PROGRAM TERMINATED BY OBJ ERROR IN CELS WITH INDIVIDUAL 1 ID= 1.00000000000000E+00 SU= M OF "SQUARED" WEIGHTED INDIVIDUAL RESIDUALS IS INFINITE MESSAGE= ISSUED FROM ESTIMATION STEP AT INITIAL OBJ. FUNCTION EVALUATION= My control file is $INPUT ID STUD STYP DOSE AMT EVID= MDV TIME DV HEIG WT BMI AGE SEX RACE $DATA ../data/M_PK.NM.csv = IGNORE=# $SUBROUTINE ADVAN6 TRANS1 TOL=3$MODEL NCOMP=3= COMP=(DEPOT DEFDOSE)COMP=(CENT DEFOBS)COM= P=(PERIPH
subject ] [ by author ] [ by messages with attachments ] From: Yuhong Chen Date: Tue, 14 http://www.cognigencorp.com/nonmem/current/2012-February/3112.html Feb 2012 12:08:58 -0500 Thanks a lot for the suggestions, it is working https://nonmem.iconplc.com/nonmem/Nonmem_7_Project/nmvi2.0beta/htmlplus/nonerr.htm now. Yuhong On Mon, Feb 13, 2012 at 5:00 PM, Leonid Gibiansky wrote: > Yuhong, > Usually this error points out to zero variance of the residual error. In > this case, you have proportional error model. If prediction is zero (or > close to zero) at some point, error in you may get it. I would try to estimate > (rather than fix to zero) the additive part of the error model. One > possible explanation is ALAG1 use: if the first observation is before ALAG1 > time, you will get zero prediction (and this error) > Thanks > Leonid > > > ------------------------------**-------- > Leonid Gibiansky, Ph.D. > President, QuantPharm LLC > web: error in cels www.quantpharm.com > e-mail: LGibiansky at quantpharm.com > tel: (301) 767 5566 > > > > > On 2/13/2012 4:20 PM, Yuhong Chen wrote: > >> Dear nmusers, >> I have a compound with nonlinear CL. I tried to run the nonlinear model >> but got error message as following, could someone point out what could >> be the error in my control file or my data? >> Best regards, >> Yuhong >> >> 0PROGRAM TERMINATED BY OBJ >> >> ERROR IN CELS WITH INDIVIDUAL 1 ID= 1.00000000000000E+00 >> >> SUM OF "SQUARED" WEIGHTED INDIVIDUAL RESIDUALS IS INFINITE >> >> MESSAGE ISSUED FROM ESTIMATION STEP >> >> AT INITIAL OBJ. FUNCTION EVALUATION >> >> My control file is >> >> $INPUT ID STUD STYP DOSE AMT EVID MDV TIME DV HEIG WT BMI AGE SEX RACE >> >> $DATA ../data/M_PK.NM.csv IGNORE=# >> >> $SUBROUTINE ADVAN6 TRANS1 TOL=3 >> $MODEL NCOMP=3 >> COMP=(DEPOT DEFDOSE) >> COMP=(CENT DEFOBS) >> COMP=(PERIPH) >> >> $PK >> ALAG1 = THETA(1) >> KA = THETA(2) >> V2 = THETA(3) >> Q = THETA(4) >> V3 = THETA(5) >> TVM = THETA(6) >> KM = THETA(7) >> >> ;################